Space-saving high-density modular data center and an energy-efficient cooling system
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional data center cooling systems are inefficient and costly due to their large, oversized designs, which lead to high energy consumption and limited geographical applicability, especially in high-density data centers with fluctuating IT loads and extreme wet bulb conditions.
Innovation Solution
A modular data center system featuring a close-coupled cooling system with a primary cooling device, heat exchange members, and a secondary cooling device that uses a combination of free-cooling and mechanical sub-cooling to efficiently manage cooling in high-density environments, utilizing polygonal shapes for space-saving and energy-efficient server rack configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a traditional chiller plant is designed to cool the entire data center, then the cooling coverage is comprehensive, but the energy consumption increases significantly and the system efficiency drops when the data center does not run at full capacity
Solution Approach 1:
The data center is divided into multiple zones or sections, each with its own cooling system. This allows only the necessary areas to be cooled at any given time, matching cooling capacity to actual IT load distribution and avoiding energy waste in unused areas.
Solution Approach 2:
The cooling system is designed to dynamically adjust its capacity and operation based on real-time IT load conditions. When load fluctuates, the system scales cooling output accordingly, maintaining reliable coverage where needed while reducing energy consumption in areas with lower demands.
2Ease of manufacture
If an air-cooled free cooling system is used, then the initial capital cost is reduced, but the system operates only in cool, dry-climate environments thereby restricting its use to limited geographical areas
Solution Approach 1:
An adiabatic cooling system is introduced as an intermediary between the air-cooled free cooling system and the IT equipment. The adiabatic process uses water evaporation to pre-cool the air before it enters the free cooling system, enabling the system to operate effectively in warmer and more humid climates where direct air cooling would fail.
Solution Approach 2:
The system changes the thermodynamic parameters of the cooling air by using adiabatic cooling to reduce air temperature and increase humidity before it reaches the free cooling exchangers. This parameter transformation allows the system to function in geographical areas with higher ambient temperatures and humidity levels.
3Adaptability or versatility
If an adiabatic-assisted cooling system is used, then the geographical reach is expanded, but the system has cooling tolerance limitations and is incapable of providing sufficient cooling to high density data centers
Solution Approach 1:
The system merges adiabatic cooling with mechanical refrigeration in a hybrid architecture. The adiabatic system provides base cooling and handles the majority of the thermal load, while mechanical refrigeration units provide supplemental cooling capacity specifically for high-density areas where heat generation exceeds adiabatic cooling capabilities.
Solution Approach 2:
Different cooling strategies are applied to different areas of the data center based on local heat density. High-density server racks receive enhanced mechanical cooling, while lower-density areas rely on adiabatic-assisted free cooling, optimizing both geographical adaptability and cooling capacity where needed.
4Reliability
If a traditional oversized cooling infrastructure is used, then the system can handle peak loads, but the initial capital cost and operation and maintenance costs are high
Solution Approach 1:
The cooling infrastructure is segmented into modular units that can be deployed incrementally. Each module provides a specific cooling capacity, and multiple modules can be combined to handle peak loads. This modular approach reduces initial capital cost by allowing phased deployment while maintaining the ability to scale up to meet peak demand requirements.
Solution Approach 2:
The cooling system is designed with multi-functionality to handle both base loads and peak loads using the same infrastructure. The adiabatic-assisted free cooling system provides economical base cooling, while mechanical refrigeration units provide peak cooling capacity, eliminating the need for separate oversized infrastructure dedicated solely to peak load handling.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The modular system significantly reduces energy costs and operational expenses by efficiently cooling high-density data centers, even in areas with high wet bulb temperatures, while requiring less mechanical refrigeration capacity and allowing for scalable deployment.
Implementation Method 1
a free-cooling system configured to cool a first fluid in thermal communication with electronic equipment using atmospheric air
Implementation Method 2
a mechanical sub-cooling system coupled to the free-cooling system, the mechanical system configured to cool a second fluid flowing in the free-cooling system as a function of an amount by which the free-cooling system has exceeded its maximum cooling capacity
Implementation Method 3
heat exchange members, and a secondary cooling device that uses a combination of free-cooling and mechanical sub-cooling to efficiently manage cooling in high-density environments
Data Source
Figure 1
Figure 2A~2B
Figure 2C~2D
AI summary
A space-saving, high-density modular data pod system and an energy-efficient cooling system are disclosed. The modular data pod system includes a central free-cooling system and a plurality of modular data pods, each of which includes a heat exchange assembly coupled to the central free-cooling system, and a distributed mechanical cooling system coupled to the heat exchange assembly. The modular data pods include a data enclosure having at least five walls arranged in the shape of a polygon, a plurality of computer racks arranged in a circular or U-shaped pattern, and a cover to create hot and cold aisles, and an air circulator configured to continuously circulate air between the hot and cold aisles. Each modular data pod also includes an auxiliary enclosure containing a shared fluid and electrical circuit section that is configured to connect to adjacent shared fluid and electrical circuit sections to form a shared fluid and electrical circuit that connects to the central free-cooling system. The auxiliary enclosure contains at least a portion of the distributed mechanical cooling system, which is configured to trim the cooling performed by the central free-cooling system.